CN111402212B - Extraction method of dynamic connection activity mode of sea person brain function network - Google Patents
Extraction method of dynamic connection activity mode of sea person brain function network Download PDFInfo
- Publication number
- CN111402212B CN111402212B CN202010144112.XA CN202010144112A CN111402212B CN 111402212 B CN111402212 B CN 111402212B CN 202010144112 A CN202010144112 A CN 202010144112A CN 111402212 B CN111402212 B CN 111402212B
- Authority
- CN
- China
- Prior art keywords
- sea
- equal
- tested
- dynamic
- function connection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000003925 brain function Effects 0.000 title claims abstract description 63
- 230000000694 effects Effects 0.000 title claims abstract description 12
- 238000000605 extraction Methods 0.000 title claims description 11
- 230000006870 function Effects 0.000 claims abstract description 77
- 238000000034 method Methods 0.000 claims abstract description 50
- 239000013598 vector Substances 0.000 claims abstract description 39
- 239000011159 matrix material Substances 0.000 claims abstract description 34
- 210000004556 brain Anatomy 0.000 claims abstract description 21
- 230000000284 resting effect Effects 0.000 claims abstract description 21
- 238000002599 functional magnetic resonance imaging Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 23
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 9
- 238000012880 independent component analysis Methods 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 5
- KJONHKAYOJNZEC-UHFFFAOYSA-N nitrazepam Chemical compound C12=CC([N+](=O)[O-])=CC=C2NC(=O)CN=C1C1=CC=CC=C1 KJONHKAYOJNZEC-UHFFFAOYSA-N 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 230000004886 head movement Effects 0.000 claims description 3
- 239000000470 constituent Substances 0.000 claims description 2
- 238000010219 correlation analysis Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims description 2
- 238000012887 quadratic function Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 7
- 230000001537 neural effect Effects 0.000 abstract description 5
- 238000011160 research Methods 0.000 abstract description 4
- 238000011156 evaluation Methods 0.000 description 3
- 230000009323 psychological health Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000004630 mental health Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241000764238 Isis Species 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008035 nerve activity Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000008925 spontaneous activity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000003936 working memory Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Algebra (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses a method for extracting a dynamic connection activity mode of a brain function network of a sea person, which comprises the following steps: step 1: collecting brain resting state functional magnetic resonance imaging data of sea men tested and non-sea men tested; step 2: preprocessing the acquired data; step 3: obtaining resting brain function networks of a plurality of groups of levels and individual levels and corresponding time courses thereof; step 4: calculating a dynamic function connection matrix and a corresponding dynamic function connection vector between each tested corresponding brain function network in the sea member and non-sea member data; step 5: extracting the sea-person specific brain function connection mode hidden in the dynamic function connection matrix from the dynamic function connection vector. The invention is helpful to acquire the specific brain function connection mode of the sea person professional group according to the dynamics; the dynamic brain function connection mode of the sea staff extracted by the invention can provide a basis for further research and analysis for the neural activity rule of the sea staff and the professional brain plasticity.
Description
Technical Field
The invention relates to the technical field of medical imaging image processing, in particular to an extraction method of a dynamic connection activity mode of a brain function network of a sea person.
Background
As a special professional group, marine working conditions faced by sea crews are greatly different from those of land environments, and are easily influenced by a plurality of complex factors such as natural environments, working environments and the like, so that psychological defects of the sea crews are caused. The mental health bad condition not only affects the physical and psychological health of seamen, but also makes the navigation operation face great potential safety hazard. It is therefore important to discover and psychologically dredge sea men in psychological sub-health conditions in advance. In recent years, the psychological health of sea crews has received increasing attention from society, especially the shipping industry. However, to our knowledge, there are few objective quantitative methods to assess the mental health of sea personnel. The traditional marine psychological health evaluation system mainly adopts a questionnaire manner, such as a symptom self-evaluation scale and the like, and the manner is easily influenced by the imperfection of the questionnaire design and subjectivity of an evaluator to be tested in answering questions, so that the evaluation result is inaccurate.
In many medical image analysis technologies, functional magnetic resonance imaging is a method for revealing brain nerve activity from a functional perspective, has the advantages of no invasiveness, no radioactivity, higher spatial and temporal resolution and the like, and is particularly widely applied in clinic based on blood oxygen level dependence. The resting state functional magnetic resonance imaging can study resting state brain functional connection through spontaneous activities of neurons, so that the resting state functional magnetic resonance imaging method is more suitable for revealing brain functional neural activity rules of sea-person groups.
However, current research is mainly focused on the aspect of static functional connection, but the brain is a complex structure, the functional connection between different brain regions changes dynamically along with time, and for special professional groups, the dynamic change abnormality often appears in the abnormality of the corresponding brain functional connection. The invention aims at extracting the special dynamic brain function connection mode of the sea-person professional group based on dynamic function connection through a certain algorithm, and further researching the characteristic of the brain function connection mode of the sea-person on the basis, thereby providing a basis for exploring the brain plasticity and the neural activity specificity of the sea-person professional group.
Disclosure of Invention
The invention aims to provide an extraction method of a dynamic connection activity mode of a sea-person brain function network, which is characterized in that a plurality of resting brain function networks and corresponding time processes thereof in sea-person function magnetic resonance data are extracted through a time cascading group independent component analysis method, then a time sliding window method is used for calculating a dynamic function connection matrix and corresponding dynamic function connection vectors between the brain function networks corresponding to each tested, finally an affine propagation clustering algorithm is used for carrying out clustering analysis on all dynamic function connection vector sets, and a specific dynamic brain function connection mode of sea-person is extracted, so that a data basis is provided for subsequent further analysis.
In order to achieve the above purpose, the invention provides a method for extracting dynamic connection activity modes of a brain function network of a sea person, which comprises the following steps:
step 1: collecting brain resting state functional magnetic resonance imaging data of sea men tested and non-sea men tested;
step 2: preprocessing acquired sea man and non-sea man resting state functional magnetic resonance imaging data, wherein the preprocessing operation comprises four steps of time layer correction, head movement correction, space standardization and space smoothing;
step 3: according to the pretreated resting state functional magnetic resonance imaging data of sea men and non-sea men, a plurality of groups of resting state brain functional networks of horizontal and individual horizontal and corresponding time processes are respectively obtained by using a time cascading group independent component analysis method and a space-time double regression mode;
step 4: calculating a dynamic function connection matrix and a corresponding dynamic function connection vector between each tested corresponding brain function network in the sea member and non-sea member data by using a sliding time window method;
step 5: extracting the sea member specific brain function connection mode hidden in the dynamic function connection matrix from the dynamic function connection vector by utilizing an affine propagation clustering algorithm.
The extraction method of the dynamic connection activity mode of the sea person brain function network comprises the following steps:
step 3.1, assuming that the group data contains K tested objects, each tested object contains T time points and V voxels after being preprocessed; and (3) performing independent component analysis of the tested level of the group by adopting a time cascading mode to obtain the following model:
(X 1 ;X 2 ;…;X K )=MS (1)
where M represents kt×v-order group mixing matrix, s= (S) 1 ,s 2 ,…,s N ) ' represents a source signal matrix of order N x V, each row representing a constituent; n is the number of brain function networks corresponding to each tested;
solving the model in a constraint optimization mode:
maximization: j(s) i )={E[G(s i )]-E[G(v)]} 2 (2)
The constraint is as follows: h(s) i )=E[s i ] 2 -1=0
Wherein s is i Represents the output component, J(s) i ) A comparison function representing the independence of the metric output components; e (·) represents the desired operation; g (·) is a non-quadratic function and v is a Gaussian random variable; equation constraint h(s) i ) The optimization problem is solved in the convex area;
step 3.2, selecting a resting brain function network of interest and a time course thereof from the composition components obtained in the step 3.1, and obtaining a brain function network corresponding to each tested in the group and a time course thereof by a space-time double regression mode; for test i (i=1, 2,., K), the expression is as follows:
M i =X i pinv(S),S i =pinv(M i )X i (3)
wherein X is i Representing an observation data matrix of order T x V, M i Representing T×N i A hybrid matrix of the order is provided,represents N i Source signal matrix of order x V, each row representing an independent component of test i,/-, is shown>Is a column vector of size V x 1.
The extraction method of the dynamic connection activity mode of the sea person brain function network comprises the following steps:
step 4.1, obtaining N brain function networks corresponding to each tested and time course T thereof through the step 3 1 、T 2 ……T N Sliding time window correlation analysis method is adopted, window width is W, step length is 1, sliding is performed on time course, and time course of nth brain function network under jth time window is recorded as(N is not less than 1 and not more than N; j is not less than 1 and not more than T-W+1); then, the Person correlation coefficient between the brain function network time processes corresponding to the tested is calculated to obtain T-W+1 dynamic function connection matrixes which form a tested dynamic function connection matrix set DFCMS= { DFCM 1 ,DFCM 2 ,...,DFCM j ,...,DFCM T-W+1 };
The Pelson correlation coefficient between every two brain function network time processes refers to the x and y time processes of the brain function network under the jth sliding windowAnd->The pearson correlation coefficient between the two is as follows:
in the method, in the process of the invention,is->And->Covariance of->Respectively isThe variance of (1) is equal to or less than or equal to j and is equal to or less than or equal to T-W+1, x is equal to or less than or equal to 1 and is equal to or less than or equal to N, and y is equal to or less than or equal to 1 and is equal to or less than or equal to N;
dynamic function connection matrix DFCM j The dynamic function connection matrix composed of pearson correlation coefficients between every two brain function network time processes under the jth sliding window is specifically expressed as:
wherein j is more than or equal to 1 and less than or equal to T-W+1, u is more than or equal to 1 and less than or equal to N, v is more than or equal to 1 and less than or equal to N;
step 4.2, for each tested, calculating a dynamic function connection vector set, wherein the specific method is as follows: the dynamic function connection matrix DFCM in the dynamic function connection matrix set DFCMS j (1. Ltoreq.j. Ltoreq.T-W+1), DFCM is arranged in rows j The upper triangle elements are spread into a row to obtain a dynamic function connection vector DFCV j (1.ltoreq.j.ltoreq.T-W+1); each column vector has a size ofThe T-W+1 column vectors are cascaded from small to large according to window time points to form a dynamic function connection vector set DFCVS= [ DFCV ] 1 ,DFCV 2 ,…,DFCV j ,…,DFCV M-W+1 ]The size is (T-W+1) multiplied by N;
wherein the DFCV j The dynamic function connection vector under the j-th sliding window is specifically expressed as:
wherein j is more than or equal to 1 and less than or equal to T-W+1, u is more than or equal to 1 and less than or equal to N, v is more than or equal to 1 and less than or equal to N.
The extraction method of the dynamic connection activity mode of the sea person brain function network comprises the following steps:
step 5.1, merging all tested dynamic function connection vector sets according to columns to form clustered samples, wherein each sample is a tested corresponding dynamic function connection vector; q initial class cores are generated by adopting a method based on an automatic target generation process;
step 5.2, clustering all tested dynamic function connection vector samples by adopting an affine propagation clustering algorithm according to the initial class core obtained in the previous step to obtain Q classes;
step 5.3, respectively calculating the corresponding relation between the respective dynamic function connection modes of the sea person tested group and the non-sea person control group, wherein the corresponding relation is expressed as the maximum correlation coefficient between the dynamic function connection modes; then analyzing the difference and the specificity between the corresponding dynamic function modes;
and 5.4, for each category in the sea person tested and non-sea person control group, calculating the number of the dynamic function connection matrixes belonging to the category in each tested, analyzing the difference and the specificity between the corresponding dynamic function connection modes, and deducing the specific dynamic brain function connection mode of the sea person tested.
Compared with the prior art, the invention has the following beneficial effects:
the method for introducing dynamic function connection is beneficial to acquiring the specific brain function connection mode of the sea person professional group according to the dynamics; the accuracy and efficiency of dynamic brain function connection mode extraction are improved by combining the group independent component analysis, the sliding time window correlation, the affine propagation clustering and the like; the dynamic brain function connection mode of the sea staff extracted by the invention can provide a basis for further research and analysis for the neural activity rule of the sea staff and the professional brain plasticity.
Drawings
FIG. 1 is a flow chart of the extraction method of the dynamic connection activity mode of the brain function network of the sea man.
Detailed Description
The invention is further described by the following examples, which are given by way of illustration only and are not limiting of the scope of the invention.
As shown in fig. 1, a method for extracting dynamic connection activity modes of a brain function network of a sea person comprises the following steps:
step 1, respectively collecting brain resting state functional magnetic resonance data of a sea person tested group and a normal non-sea person control group, wherein the number of the two groups of samples is 88, and the total number of the two groups of samples is 176. The test is required to keep the brain awake in the data acquisition process and lie in the magnetic resonance instrument. The number of time points for each of the functional magnetic resonance data tested is 215.
And step 2, preprocessing the acquired two groups of resting-state functional magnetic resonance data, wherein the preprocessing comprises four steps of time layer correction, head movement correction, spatial standardization and spatial smoothing. The preprocessing of all data is accomplished by DPARSF software.
And step 3, obtaining a dynamic function connection matrix and a dynamic function connection vector corresponding to each tested by adopting a mode of group independent component analysis and sliding time window analysis according to the preprocessed resting state functional magnetic resonance data.
Step 3.1, calculating the interested brain function network and the time course of each tested brain function network in the group, wherein the specific method is as follows: according to the preprocessed resting state functional magnetic resonance data, using time-cascaded group independent component analysis and calculation to obtain interested nine resting state brain functional networks and time sequences thereof, wherein the interested nine resting state brain functional networks comprise a default network, a visual network, a two-side visual network, an auditory network, a sensory-motor network, an execution control network, a highlight network, a working memory network and an attention network, and obtaining brain functional networks corresponding to each tested in the group and time process information thereof in a time-space double regression mode, wherein the length of the time process is M=215 TRs.
Step 3.2, calculating a dynamic function connection matrix set between nine brain function networks corresponding to each tested in the group, wherein the specific method is as follows: adopting a sliding window method, wherein the window width is 20TRs, the step length is 1TR, and the window width is in the corresponding time process T of nine brain function networks 1 、T 2 ……T 9 Upper sliding and recording the time course of the nth brain function network under the jth time window as(1.ltoreq.n.ltoreq. 9;1.ltoreq.j.ltoreq.T-W+1=196). Then, the Pelson correlation coefficients between the nine brain function network time processes corresponding to the tested are calculated to obtain 196 dynamic function connection matrixes which form a tested dynamic function connection matrix set DFCMS= { DFCM 1 ,DFCM 2 ,...,DFCM j ,...,DFCM 196 }。
Further, the pearson correlation coefficient between every two brain function network time processes specifically refers to the j-th sliding window brain function network X and y time processesAnd->The pearson correlation coefficient between the two is as follows:
in the method, in the process of the invention,is->And->Covariance of->Respectively isIs not less than 1 but not more than 196,1 but not more than x but not more than 9, and is not less than 1 but not more than y but not more than 9.
Further, dynamic function connection matrix DFCM j Refers to the dynamic function composed of Pelson correlation coefficients between every two of all brain function network time processes under the jth sliding windowThe connection matrix is specifically expressed as:
wherein j is more than or equal to 1 and less than or equal to 196,1, u is more than or equal to 9, v is more than or equal to 1 and less than or equal to 9.
Step 3.3, calculating a dynamic function connection vector set between nine brain function networks corresponding to each tested in the group, wherein the specific method is as follows: the dynamic function connection matrix DFCM in the dynamic function connection matrix set DFCMS j (1. Ltoreq.j. Ltoreq.196) DFCM by row j The upper triangle elements are spread into a row to obtain a dynamic function connection vector DFCV j (1.ltoreq.j.ltoreq.196); each column vector has a size of 36×1; the 196 column vectors are cascaded from small to large according to window time points to form a dynamic function connection vector set DFCVS= [ DFCV ] 1 ,DFCV 2 ,…,DFCV j ,…,DFCV 196 ]The size is 196×36.
Wherein the DFCV j The dynamic function connection vector under the j-th sliding window is specifically expressed as:
similarly, 1.ltoreq.j.ltoreq. 196,1.ltoreq.u.ltoreq.9, 1.ltoreq.v.ltoreq.9.
And 4, carrying out cluster analysis on all tested dynamic function connection vector sets by using an affine propagation clustering algorithm.
And 4.1, respectively merging two groups of tested dynamic function connection vector sets according to columns to form clustered samples, wherein each sample is a tested corresponding dynamic function connection column vector.
And 4.2, clustering each group of tested dynamic function connection strength column vector samples by using an affine propagation clustering algorithm to obtain 4 and 6 categories respectively.
And step 5, extracting dynamic brain function connection modes corresponding to the sea member groups according to the clustering analysis results.
And 5.1, respectively calculating the corresponding relation between the respective dynamic function connection modes of the tested sea member and the non-sea member control group, wherein the corresponding relation is expressed as the maximum correlation coefficient between the dynamic function connection modes. The differences and specificities between the corresponding dynamic functional modes are then analyzed.
And 5.2, for each category in the sea person tested and non-sea person control group, calculating the number of the dynamic function connection matrixes belonging to the category in each tested, analyzing the difference and the specificity between the corresponding dynamic function connection modes, and deducing the specific dynamic brain function connection mode of the sea person tested.
In summary, the method for introducing dynamic functional connection is helpful to obtain the specific brain functional connection mode of the sea-person professional group according to the dynamics; the accuracy and efficiency of dynamic brain function connection mode extraction are improved by combining the group independent component analysis, the sliding time window correlation, the affine propagation clustering and the like; the dynamic brain function connection mode of the sea staff extracted by the invention can provide a basis for further research and analysis for the neural activity rule of the sea staff and the professional brain plasticity.
While the present invention has been described in detail through the foregoing description of the preferred embodiment, it should be understood that the foregoing description is not to be considered as limiting the invention. Many modifications and substitutions of the present invention will become apparent to those of ordinary skill in the art upon reading the foregoing. Accordingly, the scope of the invention should be limited only by the attached claims.
Claims (1)
1. The extraction method of the dynamic connection activity mode of the brain function network of the sea person is characterized by comprising the following steps of:
step 1: collecting brain resting state functional magnetic resonance imaging data of sea men tested and non-sea men tested;
step 2: preprocessing acquired sea man and non-sea man resting state functional magnetic resonance imaging data, wherein the preprocessing operation comprises four steps of time layer correction, head movement correction, space standardization and space smoothing;
step 3: according to the pretreated resting state functional magnetic resonance imaging data of sea men and non-sea men, a plurality of groups of resting state brain functional networks of horizontal and individual horizontal and corresponding time processes are respectively obtained by using a time cascading group independent component analysis method and a space-time double regression mode;
step 4: calculating a dynamic function connection matrix and a corresponding dynamic function connection vector between each tested corresponding brain function network in the sea member and non-sea member data by using a sliding time window method;
step 5: extracting a sea member specific brain function connection mode hidden in the dynamic function connection matrix from the dynamic function connection vector by utilizing an affine propagation clustering algorithm;
the step 3 comprises the following steps:
step 3.1, assuming that the group data contains K tested objects, each tested object contains T time points and V voxels after being preprocessed; and (3) performing independent component analysis of the tested level of the group by adopting a time cascading mode to obtain the following model:
(X 1 ;X 2 ;…;X K )=MS (1)
where M represents kt×v-order group mixing matrix, s= (S) 1 ,s 2 ,…,s N ) ' represents a source signal matrix of order N x V, each row representing a constituent; n is the number of brain function networks corresponding to each tested;
solving the model in a constraint optimization mode:
maximization: j(s) i )={E[G(s i )]-E[G(v)]} 2 (2)
The constraint is as follows: h(s) i )=E[s i ] 2 -1=0
Wherein S is i Represents the output component, J (S) i ) A comparison function representing the independence of the metric output components; e (·) represents the desired operation; g (·) is a non-quadratic function and v is a Gaussian random variable; equation constraint h (S i ) The optimization problem is solved in the convex area;
step 3.2, selecting a resting brain function network of interest and a time course thereof from the composition components obtained in the step 3.1, and obtaining a brain function network corresponding to each tested in the group and a time course thereof by a space-time double regression mode; for test i (i=1, 2, …, K), the following is expressed:
M i =X i pinv(S),S i =pinv(M i )X i (3)
wherein X is i Representing an observation data matrix of order T x V, M i Representing T×N i A hybrid matrix of the order is provided,represents N i Source signal matrix of order x V, each row representing an independent component of test i,/-, is shown>Is a column vector of size V x 1;
the step 4 comprises the following steps:
step 4.1, obtaining N brain function networks corresponding to each tested and time course T thereof through the step 3 1 、T 2 ……T N Sliding time window correlation analysis method is adopted, window width is W, step length is 1, sliding is performed on time course, and time course of nth brain function network under jth time window is recorded as DT j n (N is not less than 1 and not more than N; j is not less than 1 and not more than T-W+1); then, the Person correlation coefficient between the brain function network time processes corresponding to the tested is calculated to obtain T-W+1 dynamic function connection matrixes which form a tested dynamic function connection matrix set DFCMS= { DFCM 1 ,DFCM 2 ,...,DFCM j ,...DFCM T-W+1 };
The Pelson correlation coefficient between every two brain function network time processes refers to the x and y time processes DT of the brain function network under the jth sliding window j x And DT (DT) j y The pearson correlation coefficient between the two is as follows:
in the method, in the process of the invention,for DT j x And DT (DT) j y Covariance of (v), var (DT) j x )、var(DT j y ) DT respectively j x 、DT j y The variance of (1) is equal to or less than or equal to j and is equal to or less than or equal to T-W+1, x is equal to or less than or equal to 1 and is equal to or less than or equal to N, and y is equal to or less than or equal to 1 and is equal to or less than or equal to N;
dynamic function connection matrix DFCM j The dynamic function connection matrix composed of pearson correlation coefficients between every two brain function network time processes under the jth sliding window is specifically expressed as:
wherein j is more than or equal to 1 and less than or equal to T-W+1, u is more than or equal to 1 and less than or equal to N, v is more than or equal to 1 and less than or equal to N;
step 4.2, for each tested, calculating a dynamic function connection vector set, wherein the specific method is as follows: the dynamic function connection matrix DFCM in the dynamic function connection matrix set DFCMS j (1. Ltoreq.j. Ltoreq.T-W+1), DFCM is arranged in rows j The upper triangle elements are spread into a row to obtain a dynamic function connection vector DFCV j (1.ltoreq.j.ltoreq.T-W+1); each column vector has a size ofThe T-W+1 column vectors are cascaded from small to large according to window time points to form a dynamic function connection vector set DFCVS= [ DFCV ] 1 ,DFCV 2 ,…,DFCV j ,…,DFCV M-W+1 ]The size is (T-W+1) multiplied by N;
wherein the DFCV j The dynamic function connection vector under the j-th sliding window is specifically expressed as:
wherein j is more than or equal to 1 and less than or equal to T-W+1, u is more than or equal to 1 and less than or equal to N, v is more than or equal to 1 and less than or equal to N;
the step 5 comprises the following steps:
step 5.1, merging all tested dynamic function connection vector sets according to columns to form clustered samples, wherein each sample is a tested corresponding dynamic function connection vector; q initial class cores are generated by adopting a method based on an automatic target generation process;
step 5.2, clustering all tested dynamic function connection vector samples by adopting an affine propagation clustering algorithm according to the initial class core obtained in the previous step to obtain Q classes;
step 5.3, respectively calculating the corresponding relation between the respective dynamic function connection modes of the sea person tested group and the non-sea person control group, wherein the corresponding relation is expressed as the maximum correlation coefficient between the dynamic function connection modes; then analyzing the difference and the specificity between the corresponding dynamic function modes;
and 5.4, for each category in the sea person tested and non-sea person control group, calculating the number of the dynamic function connection matrixes belonging to the category in each tested, analyzing the difference and the specificity between the corresponding dynamic function connection modes, and deducing the specific dynamic brain function connection mode of the sea person tested.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010144112.XA CN111402212B (en) | 2020-03-04 | 2020-03-04 | Extraction method of dynamic connection activity mode of sea person brain function network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010144112.XA CN111402212B (en) | 2020-03-04 | 2020-03-04 | Extraction method of dynamic connection activity mode of sea person brain function network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111402212A CN111402212A (en) | 2020-07-10 |
CN111402212B true CN111402212B (en) | 2023-11-14 |
Family
ID=71428523
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010144112.XA Active CN111402212B (en) | 2020-03-04 | 2020-03-04 | Extraction method of dynamic connection activity mode of sea person brain function network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111402212B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109316188A (en) * | 2018-09-30 | 2019-02-12 | 上海海事大学 | A kind of extracting method of migraine brain function connection mode |
CN113509148B (en) * | 2021-04-28 | 2022-04-22 | 东北大学 | Schizophrenia detection system based on mixed high-order brain network |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106204581A (en) * | 2016-07-08 | 2016-12-07 | 西安交通大学 | Based PC A and the dynamic brain function connection mode decomposition method of K mean cluster |
CN109316188A (en) * | 2018-09-30 | 2019-02-12 | 上海海事大学 | A kind of extracting method of migraine brain function connection mode |
CN110232332A (en) * | 2019-05-23 | 2019-09-13 | 中国人民解放军国防科技大学 | Extraction and brain state classification method and system for dynamic function connection local linear embedded features |
CN110265148A (en) * | 2019-06-20 | 2019-09-20 | 上海海事大学 | A kind of dynamic function pattern learning method that fMRI brain network mechanism inspires |
CN110598793A (en) * | 2019-09-12 | 2019-12-20 | 常州大学 | Brain function network feature classification method |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105816173B (en) * | 2015-01-06 | 2020-09-11 | 西门子(中国)有限公司 | Method for determining intra-cortical working state and inter-cortical working state of brain functional network |
US20190231230A1 (en) * | 2018-01-30 | 2019-08-01 | Soochow University | Cerebral function state evaluation device based on brain hemoglobin information |
-
2020
- 2020-03-04 CN CN202010144112.XA patent/CN111402212B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106204581A (en) * | 2016-07-08 | 2016-12-07 | 西安交通大学 | Based PC A and the dynamic brain function connection mode decomposition method of K mean cluster |
CN109316188A (en) * | 2018-09-30 | 2019-02-12 | 上海海事大学 | A kind of extracting method of migraine brain function connection mode |
CN110232332A (en) * | 2019-05-23 | 2019-09-13 | 中国人民解放军国防科技大学 | Extraction and brain state classification method and system for dynamic function connection local linear embedded features |
CN110265148A (en) * | 2019-06-20 | 2019-09-20 | 上海海事大学 | A kind of dynamic function pattern learning method that fMRI brain network mechanism inspires |
CN110598793A (en) * | 2019-09-12 | 2019-12-20 | 常州大学 | Brain function network feature classification method |
Non-Patent Citations (2)
Title |
---|
fMRI低频振幅比在海员心理评估方面的比较分析;高剑奇等;《磁共振成像》;20160820(第08期);全文 * |
基于fMRI的人脑功能可塑性研究――以海员职业为例;高剑奇等;《中国科技论文》;20160923(第18期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111402212A (en) | 2020-07-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ahmed et al. | Single volume image generator and deep learning-based ASD classification | |
US9687199B2 (en) | Medical imaging system providing disease prognosis | |
CN109993230B (en) | TSK fuzzy system modeling method for brain function magnetic resonance image classification | |
CN113052113B (en) | Depression identification method and system based on compact convolutional neural network | |
CN111402212B (en) | Extraction method of dynamic connection activity mode of sea person brain function network | |
CN113662545B (en) | Personality assessment method based on emotion electroencephalogram signals and multitask learning | |
Dehnavi et al. | Detection and classification of cardiac ischemia using vectorcardiogram signal via neural network | |
CN117172294B (en) | Method, system, equipment and storage medium for constructing sparse brain network | |
CN112508953A (en) | Meningioma rapid segmentation qualitative method based on deep neural network | |
CN111568412A (en) | Method and device for reconstructing visual image by utilizing electroencephalogram signal | |
CN110192860B (en) | Brain imaging intelligent test analysis method and system for network information cognition | |
Struthers et al. | Bridging the pond: measuring policy positions in the United States and Europe | |
CN113255789B (en) | Video quality evaluation method based on confrontation network and multi-tested electroencephalogram signals | |
Huang et al. | Automatic recognition of schizophrenia from facial videos using 3D convolutional neural network | |
Marvin et al. | Cardiotocogram biomedical signal classification and interpretation for fetal health evaluation | |
Zhang et al. | Blood vessel segmentation in fundus images based on improved loss function | |
CN109002798B (en) | Single-lead visual evoked potential extraction method based on convolutional neural network | |
CN114464319B (en) | AMS susceptibility assessment system based on slow feature analysis and deep neural network | |
Fisher et al. | Deep learning for comprehensive forecasting of Alzheimer's Disease progression | |
Demirci et al. | Functional magnetic resonance imaging–implications for detection of schizophrenia | |
CN114266738A (en) | Longitudinal analysis method and system for mild brain injury magnetic resonance image data | |
Ahmadieh et al. | Visual image reconstruction based on EEG signals using a generative adversarial and deep fuzzy neural network | |
Adhikary | Identification of Novel Diagnostic Neuroimaging Biomarkers for Autism Spectrum Disorder Through Convolutional Neural Network-Based Analysis of Functional, Structural, and Diffusion Tensor Imaging Data Towards Enhanced Autism Diagnosis | |
Abirami et al. | Diagnosis of Tuberculosis Using Deep Learning Models | |
West et al. | Complete imputation of missing repeated categorical data: one‐sample applications |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |